The lux regulon of Vibrio fischeri consists of two divergently transcribed operons separated by a common regulatory spacer region. The translational starts of the two operons are separated by 218 base pairs. Expression of the lux system is controlled by a complex series of interactive autoregulatory loops capable of maintaining >105-fold differences in the levels of bioluminescence in induced and uninduced cultures. While there are analogous regulatory systems in procaryotes, the lux regulon is, to our knowledge, the only regulatory system in which an end-product causes positive feedback via induced transcriptional activation. The primary regulation involves a positive regulatory protein, LuxR, the sole gene product of the left operon, and a small molecule effector, N-(3-oxohexanoyl)homoserine lactone, or "autoinducer". The autoinducer is produced from cytoplasmic precursors by the Luxl gene product, the first gene of the right operon. Autoinducer is freely diffusible and equilibrates with the growth medium. Transcription of operonL requires CAP-cAMP; if autoinducer ins present (i.e., if the cell density is sufficiently high to allow a high local concentration of autoinducer), the LuxR:autoinducer complex stimulates transcription of the operonR, which results in increased autoinducer synthesis as well as a rapid production of the 5 polypeptides required for bioluminescence. The positive feedback loop results in a dramatic increase in the levels of the bioluminescence enzymes in a very short time. The system appears to be further controlled by a limiting negative autoregulatory system that prevents runaway transcription of the bioluminescence enzymes. In addition to this primary regulatory network, the system is also controlled by LexA, O2 (possibly fnr), salt, and possibly by htpR (sigma 32). While the general features of the regulatory system have been established, little is known of the molecular mechanism of this elegant regulatory network. We propose to establish the molecular basis of the regulation of the lux system. The lux system from V. fischeri functions in laboratory strains of E. coli, allowing us to employ the powerful genetic systems that have been developed. Our approach will be to utilize the techniques of mutant screening, deletion mapping, and site directed mutagenesis to determine the various regions in the regulatory DNA to which proteins appear to bind. We will test and confirm these hypotheses by footprinting, toeprinting, and additional mutagenesis experiments. Equilibrium binding constants will be estimated by both gel retardation and solution titration methods utilizing spectroscopic measurements of the binding reactions. Our long-range goal is to understand the nature of the various interaction of specific proteins with the regulatory DNA. The possibility of cooperativity in binding of the various effector molecules will be investigated closely. Finally, while there is evidence of repression by cis-acting sequences upstream of the operonR. Again, this possibility will be investigated. While there are similarities between the design of the lux system and other systems in E. coli, especially the ara C-BAD, mal T-PQ, and asn C-A regulons of E. coli and the cl-cro regulon of phage Lambda which have the same configuration, there are also marked differences. Of these four systems, only ara and cl-cro have ben studied extensively. Knowledge of the molecular mechanism of the remarkable regulation of the lux system would comprise a major addition to our understanding of this of this most fundamental aspect of living systems, control of expression.